The 210Po/210Pb method to calculate particle export: Lessons learned from the results of three GEOTRACES transects

The deviation from secular equilibrium between the natural radionuclide Po-210 (half-life: 138.4 d) and its radioactive grandparent Pb-210 (half-life: 22.3 y) has been used to examine particle export from the surface ocean. Here we combine Po-210 and Pb-210 activity results from three GEOTRACES transects: two transects of the North Atlantic Ocean (GA03: 15-40 N, and GA01: 40-60 N) and one transect of the South Pacific Ocean (GP16: 10-15 S), and estimate Po-210 export fluxes at the base of the primary production zone (PPZ) by assuming steady state (SS) without advection or diffusion of the isotopes. The SS Po-210 flux was sometimes lower at basin margins than at the open-ocean stations along the transects. High SS Po-210 flux estimations derived in the North Atlantic subtropical gyre may be associated with the atmospheric deposition of Pb-210 to the surface ocean. In this paper we also question the validity of the SS assumption and discuss the influence of vertical advection and diffusion on the overall Po-210 activity balance. The SS model may have underestimated the export flux of Po-210 at margin stations in the GA03 and GP16 transects and along the GA01 cruise track. We found that upwelling in the Peruvian coastal region and near the Greenland shelf had a dramatic impact on the estimated Po-210 flux balance. Vertical diffusion had limited influence on the Po-210 export fluxes along GA03 and GA01 in the North Atlantic whereas it added Po-210 export fluxes by as much as 190% in GP16 in the Pacific, especially at the shelf stations 1 and 4. Further, analysis of the partitioning coefficient suggested the importance of small particles in the scavenging of radionuclides. This suggests it is wise to sample small particles along with large particles to determine the ratio of the concentration of particulate organic carbon (POC) to Po-210 activity (POC/Po-210) for the lower limit of POC export flux estimations. Finally, the observation of the deficit of Po-210 relative to Pb-210 activity (ziopo/ziopb < 1) in seawater concurrent with a deficit of Po-210 in particles contradicts our understanding of the conceptual Po-210 flux model, which assumes that Po-210 activity is more effectively removed from the surface ocean via particles than Pb-210 activity. While this observation deserves more attention, we propose two possible solutions: (1) the deficit of total Po-210 relative to Pb-210 activity in the surface ocean may be due to an input of Pb-210 activity instead of/concurrent with a relative removal of Po-210 activity via particle export; or (2) the particles collected may not be identical to the ones that have originally created the observed deficit in total Po-210 activity.